Models are meant to be a simplification of a real-world situation, and the goal of modeling is to detect relationships in the data that may not be obvious otherwise.  For example, you might want to model the epidemic spread of a disease or pathogen to see how certain environmental variables affect the pattern of spread (such as neighborhood density or travel routes).  This may enable you to better predict how a disease would impact a given town or region.  Another example might have to do with agriculture and how long you can expect a plot of land to produce at a certain rate given a set of variables.  This might then be compared to a real-world situation to further refine the model for a general case.

The simplest type of model is a numerical or mathematical model.  In these models, one or more mathematical functions are examined as a system with the focus being on a single variable and how it changes as the system progresses through time (see the sidebar graph).

Within archaeology, predictive modeling techniques are often used to try to determine site locations or site location likelihood.  We constructed a predictive model for the Bureau of Land Management.  Predictive modeling is used in many environmental disciplines.

Another type of modeling that has become popular is agent-based modeling.  The focus of agent-based models is how individuals (people, animals, organisms, objects) influence the systems in which they operate.

We have been involved with the conceptualization, programming, and evaluation of epidemiological models, models involving friendship networks, models of agricultural intensification, and simple numerical models.  If you or your institution would like to consult with us about the potential for applying a model to your problem, please contact us for more information.